- Email: [email protected]
Radiation sensitivity of Salmonella isolates relative to resistance to ampicillin, chloramphenicol or gentamicin
ARTICLE IN PRESS
Radiation Physics and Chemistry 75 (2006) 1080–1086 www.elsevier.com/locate/radphyschem
Radiation sensitivity of Salmonella isolates relative to resistance to ampicillin, chloramphenicol or gentamicin$ Brendan A. Niemira, Kelly A. Lonczynski, Christopher H. Sommers US Department of Agriculture, Agriculture Research Service, Eastern Regional Research Center, Food Safety Intervention Technologies Research Unit, 600 East Mermaid Lane, Wyndmoor, PA 19038, USA Received 2 November 2005; accepted 11 March 2006
Abstract Antibiotic resistance of inoculated bacteria is a commonly used selective marker. Bacteria resistant to the antibiotic nalidixic acid have been shown to have an increased sensitivity to irradiation. The purpose of this research was to screen a collection of Salmonella isolates for antibiotic resistance and determine the association, if any, of antibiotic resistance with radiation sensitivity. Twenty-four clinical isolates of Salmonella were screened for native resistance to multiple concentrations of ampicillin (Amp), chloramphenicol (Chl), or gentamicin (Gm). Test concentrations were chosen based on established clinical minimum inhibitory concentration (MIC) levels, and isolates were classified as either sensitive or resistant based on their ability to grow at or above the MIC. Salmonella cultures were grown overnight at (37 1C) in antibiotic-amended tryptic soy broth (TSB). Native resistance to Gm was observed with each of the 24 isolates (100%). Eight isolates (33%) were shown to be resistant to Amp, while seven isolates (29%) were shown to be resistant to Chl. In separate experiments, Salmonella cultures were grown overnight (37 1C) in TSB, centrifuged, and the cell pellets were re-suspended in phosphate buffer. The samples were then gamma irradiated at doses up to 1.0 kGy. The D10 values (the ionizing radiation dose required to reduce the viable number of microorganisms by 90%) were determined for the 24 isolates and they ranged from 0.181 to 0.359 kGy. No correlation was found between the D10 value of the isolate and its sensitivity or resistance to each of the three antibiotics. Resistance to Amp or Chl is suggested as appropriate resistance marker for Salmonella test strains to be used in studies of irradiation. r 2006 Elsevier Ltd. All rights reserved. Keywords: Antibiotic resistance; Irradiation; Nalidixic acid; Salmonella; D10 value; Selective marker
1. Introduction In the past two decades, the consumption of fresh fruits and vegetables in the US has increased, and the $
Mention of trade names or commercial products in this publication is solely for the purpose of providing specific information and does not imply recommendation or endorsement by the US Department of Agriculture. Corresponding author. Tel.: +1 215 836 3784; fax: +1 215 233 6405. E-mail address: [email protected] (B.A. Niemira).
geographic sources and distribution of fresh produce have expanded greatly (Tauxe et al., 1997). This increase is due, in part, to consumer demand for more ready-toeat foods that require little preparation, as well as a trend towards healthier foods. However, most produce receives minimal processing and can be a source of bacterial contamination. The demand for these fresh products has placed greater requirements on producers to ensure food safety (Blackburn and Davies, 1994). Public health officials have documented an increase in the number of produce-associated foodborne illnesses (Thayer and Rajkowski, 1999; Sivapalasingam et al.,
0969-806X/$ - see front matter r 2006 Elsevier Ltd. All rights reserved. doi:10.1016/j.radphyschem.2006.03.004
ARTICLE IN PRESS B.A. Niemira et al. / Radiation Physics and Chemistry 75 (2006) 1080–1086
2004). This increase has been attributed to new processing techniques and a more widespread distribution of the food supply. Foley et al. (2002) offered three means by which minimally processed fruits and vegetables may become contaminated: 1) minimal processing that does not provide sterility, 2) the presence of cut surfaces or damaged plant tissues that provide a portal of entry for microbial colonization, and 3) methods used to extend shelf-life that allow longer periods for microbial growth and proliferation. An alternative antimicrobial technology is the use of ionizing radiation (Niemira, 2003). Complicating the situation further is the emergence of foodborne pathogens that are antibiotic resistant. Recent reports indicate a greater antibiotic resistance than ten or fifteen years ago, and that exposure of foodborne pathogens to antibiotics in production agriculture results in the emergence of antimicrobial resistant strains (Golding and Matthews, 2004). A wide range of antibiotics is used in crop production for disease control, and in animal agriculture for therapy, prophylaxis, and growth promotion (Golding and Matthews, 2004). Outbreaks of salmonellosis in humans have been attributed to consumption of contaminated tomatoes, mustard cress, bean sprouts, cantaloupe, watermelon, and onions (Beuchat, 1995; Horby et al., 2003; Sivapalasingam et al., 2004). Studies are underway to find suitable methods to control human foodborne pathogens associated with fresh fruits, fruit juices, fresh cut vegetables or salads, sprouts and seeds (Thayer and Rajkowski, 1999). Due to the high background microflora observed on produce, studies using these products often use selective media to isolate the organisms of interest (Niemira, 2003). However, cells injured by the chosen antimicrobial treatment may not be able to grow on the selective media, leading to overestimates of the efficacy of the treatment (Ray, 1979). In order to avoid this, Blackburn and Davies (1994) developed antibiotic-resistant strains and antibiotic-amended media for the study of foodborne pathogens. Bacteria resistant to the antibiotic nalidixic acid (NalR) have been validated for use as a marker in studies of chemical interventions (Blackburn and Davies, 1994; Taormina and Beuchat, 1999). Until recently, it has not been validated for use in food irradiation studies. Niemira (2005a) demonstrated that the radiation sensitivity of three strains of Escherichia coli O157:H7 increased after being induced to be NalR. A similar increase in radiation sensitivity was subsequently observed in NalS isolates of Salmonella induced to become NalR (Niemira, 2005b). In these studies, the NalR daughter isolates were generated by unguided selection and reculturing in media with successively higher Nal concentrations. The specific molecular nature of the resistance mechanisms employed by these isolates was uncharacterized.
1081
The mode of action of Nal is disruption of DNA replication. As ionizing radiation can cause damage to DNA and interfere with DNA repair, Niemira (2005a) hypothesized that the NalR bacteria may have altered nucleic acid synthesis and/or repair systems, which make them more susceptible to ionizing radiation. Antibiotics with modes of action unrelated to nucleic acid structure and function may be more appropriate selective agents for use in irradiation studies; but to our knowledge, this area has not been explored. Ampicillin (Amp) is an antibiotic which affects cell wall synthesis by blocking peptidoglycan synthesis. Chloramphenicol (Chl) blocks protein synthesis by binding to the bacterial 50S ribosomal subunit. Gentamicin (Gm) also blocks protein synthesis, but by binding to the 30S ribosomal subunit (Gebreyes et al., 2004; Golding and Matthews, 2004; NCCLS, 2003). The objectives of this research were to 1) screen 24 produce isolates for native resistance to Amp, Chl, and/or Gm, 2) determine the radiation D10 values for the 24 isolates, 3) compare the D10 values of the antibiotic-resistant vs. antibioticsensitive isolates for each antibiotic separately and 4) determine the correlation of antibiotic resistance with sensitivity to radiation.
2. Materials and methods 2.1. Microorganisms Twenty-four clinical isolates of Salmonella from various sources (Table 1) were obtained from Dr. Ethan Solomon (USDA, ARS, ERRC, Wyndmoor, PA). They were propagated in tryptic soy broth (TSB, Difco, Detroit, MI) and maintained at 2 1C. Individual tubes of fresh TSB were inoculated before each experiment. They were grown for 16 h at 37 1C with agitation. The starting inoculum concentration was determined by serial dilution with sterile Butterfield’s phosphate buffer (BPB, Applied Research Institute, Newtown, CT). The cell concentration in overnight TSB cultures (37 1C) was determined to be 108 cfu/mL. 2.2. Radiation D10 determination Salmonella strains were grown overnight (37 1C) in 10 mL tubes of TSB. The cultures were centrifuged (5000g) for 10 min to pelletize the cells. The pelletized cells were re-suspended in 10 mL aliquots of BPB. For each strain, six tubes, one for each irradiation dose, were prepared from the re-suspended cells. The tubes were held at 2 1C until irradiation. The inoculated BPB solutions were irradiated to doses of 0.0 (control), 0.2, 0.4, 0.6, 0.8 and 1.0 kGy. The samples were irradiated using a self-contained cesium-137 gamma radiation source (Lockheed-Georgia, Marietta, GA) containing
ARTICLE IN PRESS 1082
B.A. Niemira et al. / Radiation Physics and Chemistry 75 (2006) 1080–1086
Table 1 Radiation and antibiotic resistance of Salmonella clinical isolates Isolate
Source
D10a
R2
Ampb
Gmc
Chld
S. S. S. S. S. S. S. S. S. S. S. S. S. S. S. S. S. S. S. S. S. S. S. S.
Tomato Orange juice Alfalfa seed Cantaloupe Cantaloupe Alfalfa seed Alfalfa seed Cantaloupe Cantaloupe Cantaloupe Alfalfa seed Ovine meat Octopus Cantaloupe Beef Cantaloupe Meat Cantaloupe Cantaloupe Tomato Sprouts Orange juice Cantaloupe Sprouts
0.181 0.190 0.191 0.195 0.198 0.199 0.201 0.203 0.204 0.213 0.227 0.229 0.230 0.234 0.248 0.262 0.270 0.273 0.332 0.335 0.339 0.348 0.352 0.359
0.961 0.847 0.964 0.982 0.885 0.966 0.945 0.943 0.944 0.977 0.984 0.915 0.972 0.981 0.973 0.954 0.921 0.970 0.971 0.969 0.976 0.967 0.873 0.977
s s s r s s s s s r s r r r r s r s s s s s r s
r r r r r r r r r r r r r r r r r r r r r r r r
s s s r s s s s s s r r r s r s r s s s s s r s
Montevideo gaminara F2172 worthington TX31 poona G91-1574 poona PTVS-1 mbandaka RV1DHE bredeney 3VIPHE newport 02-216 michigan poona 348 muenchen HERV2C poona 953 poona 418 mbandaka 00916-1 st. paul FSIS-039 hidalgo 02-517-2 typhimurium DT104 st. paul 02-517-2 typhimurium RO45 baildon 61-99 stanley H0558 enteritidis 15159 gaminara 02-615 anatum F4317
s, sensitivity to antibiotic; r, resistance to antibiotic. a The D10 value is defined as the radiation dose required to produce 1–log10 or 90% reduction in viable microorganisms. b Ampicillin. c Gentamicin. d Chloramphenicol.
23 137Cs pencils placed in annular array. The dose rate was 0.090 kGy/min. The dose rate was established using alanine transfer dosimeters from the National Institutes of Standards and Technology (Gaithersburg, MD). Samples were oriented with similar geometry at the center of the irradiation field for all treatments. During irradiation, the gas phase of liquid nitrogen was flushed into the sample chamber in order to maintain the temperature of 2 1C. Alanine pellets (Bruker Inc., Billarica, MA) were used for dosimetry. The pellets were read on a Bruker EMS 104 EPR analyzer and compared with a previously determined standard curve. Actual doses were within 5% of the nominal dose. After irradiation, samples were immediately serially diluted with sterile BPB using 10-fold dilutions. After dilution, 1 mL samples were taken and pour-plated with tryptic soy agar (TSA, Difco, Detroit, MI). Three pour plates per dilution were prepared, inverted and incubated at 37 1C for 24 h. Plate colonies were counted with a Spiral Biotech laser bacteria colony counter (Exotech Inc., Gaithersburg, MD).
The data for each sample were normalized against the control and plotted as the log10 reduction using the nominal doses. Each experiment was performed three times, and the data pooled. The slopes of the individual survivor curves were calculated with linear regression (SigmaPlot 5.0, SPSS Inc., Chicago, IL), and compared among all of the isolates using analysis of covariance (ANCOVA, Excel 97, Microsoft Corp., Redmond, WA). The ionizing radiation D10 value was calculated by taking the negative reciprocal of the survivor curve slope (Quattro Pro, Corel Corp., Ottawa, Ont., Canada). 2.3. Screening for antibiotic resistance The isolates were screened for resistance to Amp, Chl or Gm using test concentrations based on established clinical minimum inhibitory concentration (MIC) levels: 32 mg/mL for Amp and Chl, and 16 mg/mL for Gm (Gebreyes et al., 2004). For Amp and Chl, the test concentrations used were 0 (control), 10, 20, 30, 40, and
ARTICLE IN PRESS B.A. Niemira et al. / Radiation Physics and Chemistry 75 (2006) 1080–1086
50 mg/mL. Based on the clinical MIC level, growth at 40 mg/mL or higher resulted in classification as resistant. For Gm, the test concentrations used were 0 (control), 5, 10, 15, 20, and 25 mg/mL. Growth at 20 mg/mL or higher resulted in classification as resistant. Separate freshly prepared tubes of sterile TSB were amended with each antibiotic to the specified concentrations using a 0.22 mm-filter-sterilized stock solution. Aliquots (200 mL) of the antibiotic-amended TSB were dispensed into 96 well plates (Nunc, Roskilde, Denmark) such that each column contained one well of each concentration. Fresh overnight cultures of each of the isolates were grown in TSB as described above, and used to inoculate the antibiotic-amended wells, 10 mL per well. The plates were read at 600 nm using a Opsys MR 96 microwell plate reader (Dynex Technologies, Chantilly, VA) to establish an absorbance baseline. The plates were covered and incubated at 37 1C overnight and absorbance taken again. For each antibiotic/isolate/concentration combination, the baseline absorbance was subtracted from the post-incubation absorbance to determine the change in turbidity. The analysis was repeated for each isolate using at least four separate plates. For a given antibiotic, the data were pooled for each isolate, and the absorbance at each concentration was compared to the 0 mg/mL control using a one-way analysis of variance test (ANOVA, Dunnet, SigmaStat, version 2.0, SPSS Chicago, IL), to identify the MIC. An isolate was identified as resistant or sensitive based on the MIC. The collection of isolates was ranked by radiation D10 values and grouped into quartiles, with 6 isolates per quartile. To determine the degree of interrelationship between radiation D10 value and antibiotic resistance, a w2 test was performed (SigmaStat 5.0, SPSS Inc., Chicago, IL) on the distribution of resistant vs. sensitive isolates for each antibiotic, based on these quartile groupings.
1083
S. montevideo, S. gaminara F2172, S. worthington TX31, and S. poona G 91-1574. The second cluster includes S. poona G91-1574, S. poona PTVS-1, S. mbandaka RV1DHE, S. bredeney 3VIPHE, S. newport 02-216, S. michigan, and S. poona 348. The third cluster includes S. poona 348, S. muenchen HERV2C, S. poona 953, S. poona 418, S. mbandaka 00916-1, and S. st. paul FSIS039. The fourth cluster includes the isolates S. st. paul FSIS-039, S. hidalgo 02-517-2, S. typhimurium DT104, and S. st. paul 02-517-2. The fifth and final cluster does not overlap with any of the previous groupings. It includes S. typhimurium RO45, S. baildon 61-99, S. stanley H0558, S. enteritidis 15159, S. gaminara 02615, and S. anatum F4317. It should be noted that while the overlap between clusters is generally by nearest neighbor, isolates in the center of some clusters with larger standard errors have a broader range of statistical similarity. Examples of these are S. poona G91-1574, S. mbandaka RV1DHE, S. poona 953, and S. typhimurium DT104. Resistant isolates were observed to grow at the highest levels of antibiotic tested. Of the 24 isolates tested, 8 (33%) were resistant to Amp, 7 (29%) were resistant to Chl, and all 24 (100%) were shown to be resistant to Gm (Table 1). Isolates resistant to Amp, but not to Chl, include S. poona 348 and S. mbandaka 00916-1. One isolate (S. muenchen HERV2C) is resistant to Chl, but not to Amp. Six isolates were resistant to all the three antibiotics. They include S. poona G91-1574, S. poona 953, S. poona 418, S. st. paul FSIS-039, S. typhimurium DT104, and S. gaminara 02-615. w2 analysis shows that the resistant isolates are not significantly associated with any one quartile of radiation resistance. No correlation between antibiotic resistance and irradiation sensitivity was observed.
4. Discussion 3. Results Ionizing radiation effectively reduced the viable population of each of the isolates (Fig. 1). The radiation sensitivity was significantly variable among the various isolates (Table 1). The minimum D10 value observed was 0.181 kGy, while the maximum D10 value was 0.359 kGy, almost double that of the minimum value. An isolate’s source was not an indication of its D10 value. The D10 values are statistically similar to nearest neighbors, and form statistical clusters based on the ANCOVA results. To facilitate visualization of this clustering, the twenty-four Salmonella isolates were plotted with increasing D10 value (Fig. 2). Five major statistical clusters are observed in the data and are marked by ellipses in Fig. 2. The first cluster includes
The Salmonella isolates tested in this study had D10 values which ranged from 0.181 to 0.359 kGy. This range is consistent with previously published D10 values for Salmonella (Thayer and Rajkowski, 1999; Niemira, 2003). The statistical overlap and clustering of the D10 values is not apparently related to the foods from which the isolates were associated. The cluster with the highest D10 values contains isolates with a similar provenance as those found in the other clusters, such as isolates from cantaloupe, orange juice and tomato. A comprehensive understanding of why a given isolate may be more or less sensitive to irradiation than related isolates of the same pathogen has yet to be formulated (Thayer and Rajkowski, 1999; Niemira, 2003). It is likely that a survey of a larger number of isolates would result in a more linear progression of D10 values, with values that
ARTICLE IN PRESS 1084
B.A. Niemira et al. / Radiation Physics and Chemistry 75 (2006) 1080–1086
Fig. 1. Radiation sensitivity of twenty-four Salmonella clinical isolates in buffer. Isolates are divided in quartiles (A, B, C and D) by increasing radiation D10 values, six isolates per quartile.
bridge the gap seen among the statistical clusters observed among these 24 isolates. Antibiotic resistance among human pathogens is of increasing concern, particularly with regard to the potential for emergence of multi-drug-resistant strains (Golding and Matthews, 2004). Each of the 24 isolates examined in this study were resistant to Gm. Published surveys indicate that the prevalence of resistance to Gm can depend on the collection being examined, and can vary from year to year. Of 87 Salmonella isolates collected from turkeys in 1998–2002, 44.8% were resistant to Gm, a proportion which varied from 21.4% to 61.1%, depending on the year (Malik et al., 2003). In a survey of 390 Salmonella abattoir isolates taken from hog, beef and chicken carcasses, approximately 5% were determined to be resistant to Gm (Larkin et al., 2004). A study of 29 Salmonella isolates taken from turkey production facilities and turkey intestinal tracts showed 52% of isolates resistant to Gm (Nayak et al., 2004). These studies cite the use of antibiotics in an agricultural setting as a likely factor in
the emergence of single- or multi-drug-resistant pathogens. The 24 isolates in the present study were taken from a variety of sources, including produce, meat and juice products. A point of commonality among them is that each of these isolates was associated with a foodborne illness outbreak. For the purposes of the present study, the uniformity of resistance to Gm means that no conclusions may be drawn with regard to radiation sensitivity. The majority of the isolates which were resistant to Amp were also resistant to Chl, suggesting the possibility of a co-inheritance of resistance genes to these antibiotics, possibly on a single plasmid. Confirmation of this possibility would further investigation and characterization of the molecular basis for the antibiotic resistance exhibited by these isolates. The results presented herein lead to the conclusion that there is no relationship between radiation D10 value, and antibiotic resistance to Amp and Chl (Table 2). Antibiotic-resistant bacteria were not preferentially associated with either low- or high-D10 value isolates
ARTICLE IN PRESS B.A. Niemira et al. / Radiation Physics and Chemistry 75 (2006) 1080–1086
1085
Fig. 2. Salmonella isolates in order of increasing radiation D10 value including error bars. Statistical clusters are marked by ellipses on the figure.
Table 2 w2 results showing no correlation between D10 value and antibiotic resistance Quartile
1st 2nd 3rd 4th w2
Ampicillin
Gentamicin
Chloramphenicol
r
s
r
s
r
s
1 2 4 1
5 4 2 5
6 6 6 6
0 0 0 0
1 2 3 1
5 4 3 5
p ¼ 0:212
p ¼ 0:99
p ¼ 0:528
s, sensitivity to antibiotic; r, resistance to antibiotic.
for either Amp or Chl. The modes of action of Amp (cell wall synthesis) and Chl (protein synthesis) are not related to nucleic acid structure or function. The primary mode of action of ionizing radiation is via hydrogen and hydroxyl radical molecules resulting from the ionization of water molecules within the target organism. These radicals can disrupt membranes and interfere with the functioning of proteins, but the most significant target within the cell is DNA, where radicals are responsible for strand breakage (Niemira, 2003). The quinolone antibiotic Nal inhibits DNA synthesis in bacteria by interfering with the functioning of DNA
gyrase and topoisomerase IV (Georgopapadakou et al., 1987; Khadursky and Cozzarelli, 1998). NalR strains of E. coli O157:H7 and Salmonella have been recently shown to be more sensitive to irradiation than the NalS parent strains from which they were derived (Niemira, 2005a,b). In the E. coli O157:H7 study (Niemira, 2005a), NalR strains were also exposed to the antibiotic before, during and after irradiation with no change in survival or radiation sensitivity. The mechanism by which the induced NalR protects the cell from the antibiotic was therefore determined to be insensitive to the irradiation process. The NalR isolates used in that study were
ARTICLE IN PRESS 1086
B.A. Niemira et al. / Radiation Physics and Chemistry 75 (2006) 1080–1086
generated using unguided selection rather than by a transformation with a specific insertion of a DNA construct. The precise molecular nature of the Nal resistance in those isolates was therefore not characterized. Nevertheless, it was concluded that the significant differences in D10 values between NalR isolates and their NalS parent isolates would lead to an overestimate of the efficacy of ionizing radiation against E. coli 0157:H7 in studies using the NalR daughter isolates (Niemira, 2005a). In the present study, resistance to Amp or Chl was not associated with sensitivity to ionizing radiation. Therefore, bacterial resistance to AmpR or ChlR, rather than NalR, are suggested as more appropriate markers in irradiation studies of inoculated food products using Salmonella. Acknowledgments The authors would like to thank Dr. E. Solomon for providing Salmonella isolates, Dr. J. Buescher for valuable discussion, and Drs. L. Huang and C.-H. Liao for their thoughtful reviews of this manuscript. References Beuchat, L.R., 1995. Pathogenic microorganisms associated with fresh produce. J. Food Prot. 59, 204–216. Blackburn, C. de W., Davies, A.R., 1994. Development of antibiotic-resistant strains for the enumeration of foodborne pathogenic bacteria in stored foods. Int. J. Food Microbiol. 24, 125–136. Foley, D.M., Dufor, A., Rodriguez, L., Caporaso, F., Prakash, A., 2002. Reduction of Escherichia coli 0157:H7 in shredded iceberg lettuce by chlorination and gamma irradiation. Radiat. Phys. Chem. 63, 391–396. Gebreyes, W.A., Davies, P.R., Turkson, P., Morrow, W.E.M., Funk, J.A., Althier, C., Thakur, S., 2004. Characterization of antimicrobial-resistant phenotypes and genotypes among Salmonella enterica recovered from pigs on farms, from transport trucks, and from pigs after slaughter. J. Food Prot. 67, 698–705. Georgopapadakou, H., Dis, B.A., Angehrn, P., Wick, A., Olsen, G.L., 1987. Monocyclic and tricyclic analogs of quinolones: mechanisms of action. Antimicrob. Agents Chemother. 31, 614–616. Golding, S.S., Matthews, K.R., 2004. Intrinsic mechanism decreases susceptibility of Escherichia coli O157:H7 to multiple antibiotics. J. Food Prot. 67, 34–39. Horby, P.W., O’Brien, S.J., Adak, G.K., Graham, C., Hawker, J.I., Hunter, P., Lane, C., Lawson, A.J., Mitchell, R.T., Reacher, M.H., Threlfall, E.J., Ward, L.R., 2003. A national outbreak of multi-resistant Salmonella enterica
serovar Typhimurium definitive phage type (DT) 104 associated with consumption of lettuce. Epidemiol. Infect. 130, 169–178. Khadursky, A.B., Cozzarelli, N.R., 1998. The mechanism of inhibition of topoisomerase IV by quinolone antibacterials. J. Biol. Chem. 273, 27668–27677. Larkin, C., Poppe, C., McNab, B., McEwen, B., Mahdi, A., Odumeru, J., 2004. Antibiotic resistance of Salmonella isolated from hog, beef and chicken carcass samples from provincially inspected abattoirs in Ontario. J. Food Prot. 67, 448–455. Malik, Y.S., Olsen, K., Chander, Y., Goyal, S.M., 2003. Antimicrobial resistance in bacterial pathogens isolated from turkeys in Minnesota from 1998 to 2002. Int. J. Appl. Res. Vet. Med. 1 (3) http://www.jarvm.com/articles/ Vol1Iss3/Malik.htm (Accessed October 5, 2005). Nayak, R., Stewart, T., Wang, R.F., Lin, J., Cerniglia, C.E., Kenney, P.B., 2004. Genetic diversity and virulence gene determinants of antibiotic-resistant Salmonella isolated from preharvest turkey sources. Int. J. Food Microbiol. 91, 51–62. NCCLS (National Committee for Clinical Laboratory Standards), 2003. Methods for dilution antimicrobial susceptibility tests for bacteria that grow aerobically, approved standard, sixth ed., NCCLS document M7-A6, Wayne, PA, ISBN 1-56238-486-4. Niemira, B.A., 2003. Irradiation of fresh and minimally processed fruits, vegetables and juices. In: Novak, J.S., Sapers, G.M., Juneja, V.K. (Eds.), The Microbial Safety of Minimally Processed Foods. CRC Press, Boca Raton, FL, pp. 279–300. Niemira, B.A., 2005a. Nalidixic acid resistance increases sensitivity of Escherichia coli 0157:H7 to ionizing radiation in solution and on green leaf lettuce. J. Food Sci. 79 (2), M121–M124. Niemira, B.A., 2005b. Nalidixic acid resistance increases sensitivity of Salmonella to ionizing radiation in buffer and in orange juice. Abs P1-33. International Association for Food Protection Annual Meeting, Baltimore, MD. Ray, B., 1979. Methods to detect stressed microorganisms. J. Food Prot. 42, 346–355. Sivapalasingam, S., Friedman, C.R., Cohen, L., Tauxe, R.V., 2004. Fresh produce: a growing cause of outbreaks of foodborne illness in the United States, 1973 through 1997. J. Food Prot. 67, 2342–2353. Taormina, P.J., Beuchat, L.R., 1999. Comparison of chemical treatments to eliminate enterohemorrhagic Escherichia coli O157:H7 on alfalfa seeds. J. Food Prot. 62, 318–324. Tauxe, R., Kruse, H., Hedberg, C., Potter, M., Madden, J., Wachsmuth, K., 1997. Microbial hazards and emerging issues associated with produce: a preliminary report to the National Advisory Committee on Microbiological Criteria for Foods. J. Food Prot. 60, 1400–1408. Thayer, D.W., Rajkowski, K.T., 1999. Developments in irradiation of fresh fruits and vegetables. Food Technol. 53, 62–65.
Radiation Physics and Chemistry 75 (2006) 1080–1086 www.elsevier.com/locate/radphyschem
Radiation sensitivity of Salmonella isolates relative to resistance to ampicillin, chloramphenicol or gentamicin$ Brendan A. Niemira, Kelly A. Lonczynski, Christopher H. Sommers US Department of Agriculture, Agriculture Research Service, Eastern Regional Research Center, Food Safety Intervention Technologies Research Unit, 600 East Mermaid Lane, Wyndmoor, PA 19038, USA Received 2 November 2005; accepted 11 March 2006
Abstract Antibiotic resistance of inoculated bacteria is a commonly used selective marker. Bacteria resistant to the antibiotic nalidixic acid have been shown to have an increased sensitivity to irradiation. The purpose of this research was to screen a collection of Salmonella isolates for antibiotic resistance and determine the association, if any, of antibiotic resistance with radiation sensitivity. Twenty-four clinical isolates of Salmonella were screened for native resistance to multiple concentrations of ampicillin (Amp), chloramphenicol (Chl), or gentamicin (Gm). Test concentrations were chosen based on established clinical minimum inhibitory concentration (MIC) levels, and isolates were classified as either sensitive or resistant based on their ability to grow at or above the MIC. Salmonella cultures were grown overnight at (37 1C) in antibiotic-amended tryptic soy broth (TSB). Native resistance to Gm was observed with each of the 24 isolates (100%). Eight isolates (33%) were shown to be resistant to Amp, while seven isolates (29%) were shown to be resistant to Chl. In separate experiments, Salmonella cultures were grown overnight (37 1C) in TSB, centrifuged, and the cell pellets were re-suspended in phosphate buffer. The samples were then gamma irradiated at doses up to 1.0 kGy. The D10 values (the ionizing radiation dose required to reduce the viable number of microorganisms by 90%) were determined for the 24 isolates and they ranged from 0.181 to 0.359 kGy. No correlation was found between the D10 value of the isolate and its sensitivity or resistance to each of the three antibiotics. Resistance to Amp or Chl is suggested as appropriate resistance marker for Salmonella test strains to be used in studies of irradiation. r 2006 Elsevier Ltd. All rights reserved. Keywords: Antibiotic resistance; Irradiation; Nalidixic acid; Salmonella; D10 value; Selective marker
1. Introduction In the past two decades, the consumption of fresh fruits and vegetables in the US has increased, and the $
Mention of trade names or commercial products in this publication is solely for the purpose of providing specific information and does not imply recommendation or endorsement by the US Department of Agriculture. Corresponding author. Tel.: +1 215 836 3784; fax: +1 215 233 6405. E-mail address: [email protected] (B.A. Niemira).
geographic sources and distribution of fresh produce have expanded greatly (Tauxe et al., 1997). This increase is due, in part, to consumer demand for more ready-toeat foods that require little preparation, as well as a trend towards healthier foods. However, most produce receives minimal processing and can be a source of bacterial contamination. The demand for these fresh products has placed greater requirements on producers to ensure food safety (Blackburn and Davies, 1994). Public health officials have documented an increase in the number of produce-associated foodborne illnesses (Thayer and Rajkowski, 1999; Sivapalasingam et al.,
0969-806X/$ - see front matter r 2006 Elsevier Ltd. All rights reserved. doi:10.1016/j.radphyschem.2006.03.004
ARTICLE IN PRESS B.A. Niemira et al. / Radiation Physics and Chemistry 75 (2006) 1080–1086
2004). This increase has been attributed to new processing techniques and a more widespread distribution of the food supply. Foley et al. (2002) offered three means by which minimally processed fruits and vegetables may become contaminated: 1) minimal processing that does not provide sterility, 2) the presence of cut surfaces or damaged plant tissues that provide a portal of entry for microbial colonization, and 3) methods used to extend shelf-life that allow longer periods for microbial growth and proliferation. An alternative antimicrobial technology is the use of ionizing radiation (Niemira, 2003). Complicating the situation further is the emergence of foodborne pathogens that are antibiotic resistant. Recent reports indicate a greater antibiotic resistance than ten or fifteen years ago, and that exposure of foodborne pathogens to antibiotics in production agriculture results in the emergence of antimicrobial resistant strains (Golding and Matthews, 2004). A wide range of antibiotics is used in crop production for disease control, and in animal agriculture for therapy, prophylaxis, and growth promotion (Golding and Matthews, 2004). Outbreaks of salmonellosis in humans have been attributed to consumption of contaminated tomatoes, mustard cress, bean sprouts, cantaloupe, watermelon, and onions (Beuchat, 1995; Horby et al., 2003; Sivapalasingam et al., 2004). Studies are underway to find suitable methods to control human foodborne pathogens associated with fresh fruits, fruit juices, fresh cut vegetables or salads, sprouts and seeds (Thayer and Rajkowski, 1999). Due to the high background microflora observed on produce, studies using these products often use selective media to isolate the organisms of interest (Niemira, 2003). However, cells injured by the chosen antimicrobial treatment may not be able to grow on the selective media, leading to overestimates of the efficacy of the treatment (Ray, 1979). In order to avoid this, Blackburn and Davies (1994) developed antibiotic-resistant strains and antibiotic-amended media for the study of foodborne pathogens. Bacteria resistant to the antibiotic nalidixic acid (NalR) have been validated for use as a marker in studies of chemical interventions (Blackburn and Davies, 1994; Taormina and Beuchat, 1999). Until recently, it has not been validated for use in food irradiation studies. Niemira (2005a) demonstrated that the radiation sensitivity of three strains of Escherichia coli O157:H7 increased after being induced to be NalR. A similar increase in radiation sensitivity was subsequently observed in NalS isolates of Salmonella induced to become NalR (Niemira, 2005b). In these studies, the NalR daughter isolates were generated by unguided selection and reculturing in media with successively higher Nal concentrations. The specific molecular nature of the resistance mechanisms employed by these isolates was uncharacterized.
1081
The mode of action of Nal is disruption of DNA replication. As ionizing radiation can cause damage to DNA and interfere with DNA repair, Niemira (2005a) hypothesized that the NalR bacteria may have altered nucleic acid synthesis and/or repair systems, which make them more susceptible to ionizing radiation. Antibiotics with modes of action unrelated to nucleic acid structure and function may be more appropriate selective agents for use in irradiation studies; but to our knowledge, this area has not been explored. Ampicillin (Amp) is an antibiotic which affects cell wall synthesis by blocking peptidoglycan synthesis. Chloramphenicol (Chl) blocks protein synthesis by binding to the bacterial 50S ribosomal subunit. Gentamicin (Gm) also blocks protein synthesis, but by binding to the 30S ribosomal subunit (Gebreyes et al., 2004; Golding and Matthews, 2004; NCCLS, 2003). The objectives of this research were to 1) screen 24 produce isolates for native resistance to Amp, Chl, and/or Gm, 2) determine the radiation D10 values for the 24 isolates, 3) compare the D10 values of the antibiotic-resistant vs. antibioticsensitive isolates for each antibiotic separately and 4) determine the correlation of antibiotic resistance with sensitivity to radiation.
2. Materials and methods 2.1. Microorganisms Twenty-four clinical isolates of Salmonella from various sources (Table 1) were obtained from Dr. Ethan Solomon (USDA, ARS, ERRC, Wyndmoor, PA). They were propagated in tryptic soy broth (TSB, Difco, Detroit, MI) and maintained at 2 1C. Individual tubes of fresh TSB were inoculated before each experiment. They were grown for 16 h at 37 1C with agitation. The starting inoculum concentration was determined by serial dilution with sterile Butterfield’s phosphate buffer (BPB, Applied Research Institute, Newtown, CT). The cell concentration in overnight TSB cultures (37 1C) was determined to be 108 cfu/mL. 2.2. Radiation D10 determination Salmonella strains were grown overnight (37 1C) in 10 mL tubes of TSB. The cultures were centrifuged (5000g) for 10 min to pelletize the cells. The pelletized cells were re-suspended in 10 mL aliquots of BPB. For each strain, six tubes, one for each irradiation dose, were prepared from the re-suspended cells. The tubes were held at 2 1C until irradiation. The inoculated BPB solutions were irradiated to doses of 0.0 (control), 0.2, 0.4, 0.6, 0.8 and 1.0 kGy. The samples were irradiated using a self-contained cesium-137 gamma radiation source (Lockheed-Georgia, Marietta, GA) containing
ARTICLE IN PRESS 1082
B.A. Niemira et al. / Radiation Physics and Chemistry 75 (2006) 1080–1086
Table 1 Radiation and antibiotic resistance of Salmonella clinical isolates Isolate
Source
D10a
R2
Ampb
Gmc
Chld
S. S. S. S. S. S. S. S. S. S. S. S. S. S. S. S. S. S. S. S. S. S. S. S.
Tomato Orange juice Alfalfa seed Cantaloupe Cantaloupe Alfalfa seed Alfalfa seed Cantaloupe Cantaloupe Cantaloupe Alfalfa seed Ovine meat Octopus Cantaloupe Beef Cantaloupe Meat Cantaloupe Cantaloupe Tomato Sprouts Orange juice Cantaloupe Sprouts
0.181 0.190 0.191 0.195 0.198 0.199 0.201 0.203 0.204 0.213 0.227 0.229 0.230 0.234 0.248 0.262 0.270 0.273 0.332 0.335 0.339 0.348 0.352 0.359
0.961 0.847 0.964 0.982 0.885 0.966 0.945 0.943 0.944 0.977 0.984 0.915 0.972 0.981 0.973 0.954 0.921 0.970 0.971 0.969 0.976 0.967 0.873 0.977
s s s r s s s s s r s r r r r s r s s s s s r s
r r r r r r r r r r r r r r r r r r r r r r r r
s s s r s s s s s s r r r s r s r s s s s s r s
Montevideo gaminara F2172 worthington TX31 poona G91-1574 poona PTVS-1 mbandaka RV1DHE bredeney 3VIPHE newport 02-216 michigan poona 348 muenchen HERV2C poona 953 poona 418 mbandaka 00916-1 st. paul FSIS-039 hidalgo 02-517-2 typhimurium DT104 st. paul 02-517-2 typhimurium RO45 baildon 61-99 stanley H0558 enteritidis 15159 gaminara 02-615 anatum F4317
s, sensitivity to antibiotic; r, resistance to antibiotic. a The D10 value is defined as the radiation dose required to produce 1–log10 or 90% reduction in viable microorganisms. b Ampicillin. c Gentamicin. d Chloramphenicol.
23 137Cs pencils placed in annular array. The dose rate was 0.090 kGy/min. The dose rate was established using alanine transfer dosimeters from the National Institutes of Standards and Technology (Gaithersburg, MD). Samples were oriented with similar geometry at the center of the irradiation field for all treatments. During irradiation, the gas phase of liquid nitrogen was flushed into the sample chamber in order to maintain the temperature of 2 1C. Alanine pellets (Bruker Inc., Billarica, MA) were used for dosimetry. The pellets were read on a Bruker EMS 104 EPR analyzer and compared with a previously determined standard curve. Actual doses were within 5% of the nominal dose. After irradiation, samples were immediately serially diluted with sterile BPB using 10-fold dilutions. After dilution, 1 mL samples were taken and pour-plated with tryptic soy agar (TSA, Difco, Detroit, MI). Three pour plates per dilution were prepared, inverted and incubated at 37 1C for 24 h. Plate colonies were counted with a Spiral Biotech laser bacteria colony counter (Exotech Inc., Gaithersburg, MD).
The data for each sample were normalized against the control and plotted as the log10 reduction using the nominal doses. Each experiment was performed three times, and the data pooled. The slopes of the individual survivor curves were calculated with linear regression (SigmaPlot 5.0, SPSS Inc., Chicago, IL), and compared among all of the isolates using analysis of covariance (ANCOVA, Excel 97, Microsoft Corp., Redmond, WA). The ionizing radiation D10 value was calculated by taking the negative reciprocal of the survivor curve slope (Quattro Pro, Corel Corp., Ottawa, Ont., Canada). 2.3. Screening for antibiotic resistance The isolates were screened for resistance to Amp, Chl or Gm using test concentrations based on established clinical minimum inhibitory concentration (MIC) levels: 32 mg/mL for Amp and Chl, and 16 mg/mL for Gm (Gebreyes et al., 2004). For Amp and Chl, the test concentrations used were 0 (control), 10, 20, 30, 40, and
ARTICLE IN PRESS B.A. Niemira et al. / Radiation Physics and Chemistry 75 (2006) 1080–1086
50 mg/mL. Based on the clinical MIC level, growth at 40 mg/mL or higher resulted in classification as resistant. For Gm, the test concentrations used were 0 (control), 5, 10, 15, 20, and 25 mg/mL. Growth at 20 mg/mL or higher resulted in classification as resistant. Separate freshly prepared tubes of sterile TSB were amended with each antibiotic to the specified concentrations using a 0.22 mm-filter-sterilized stock solution. Aliquots (200 mL) of the antibiotic-amended TSB were dispensed into 96 well plates (Nunc, Roskilde, Denmark) such that each column contained one well of each concentration. Fresh overnight cultures of each of the isolates were grown in TSB as described above, and used to inoculate the antibiotic-amended wells, 10 mL per well. The plates were read at 600 nm using a Opsys MR 96 microwell plate reader (Dynex Technologies, Chantilly, VA) to establish an absorbance baseline. The plates were covered and incubated at 37 1C overnight and absorbance taken again. For each antibiotic/isolate/concentration combination, the baseline absorbance was subtracted from the post-incubation absorbance to determine the change in turbidity. The analysis was repeated for each isolate using at least four separate plates. For a given antibiotic, the data were pooled for each isolate, and the absorbance at each concentration was compared to the 0 mg/mL control using a one-way analysis of variance test (ANOVA, Dunnet, SigmaStat, version 2.0, SPSS Chicago, IL), to identify the MIC. An isolate was identified as resistant or sensitive based on the MIC. The collection of isolates was ranked by radiation D10 values and grouped into quartiles, with 6 isolates per quartile. To determine the degree of interrelationship between radiation D10 value and antibiotic resistance, a w2 test was performed (SigmaStat 5.0, SPSS Inc., Chicago, IL) on the distribution of resistant vs. sensitive isolates for each antibiotic, based on these quartile groupings.
1083
S. montevideo, S. gaminara F2172, S. worthington TX31, and S. poona G 91-1574. The second cluster includes S. poona G91-1574, S. poona PTVS-1, S. mbandaka RV1DHE, S. bredeney 3VIPHE, S. newport 02-216, S. michigan, and S. poona 348. The third cluster includes S. poona 348, S. muenchen HERV2C, S. poona 953, S. poona 418, S. mbandaka 00916-1, and S. st. paul FSIS039. The fourth cluster includes the isolates S. st. paul FSIS-039, S. hidalgo 02-517-2, S. typhimurium DT104, and S. st. paul 02-517-2. The fifth and final cluster does not overlap with any of the previous groupings. It includes S. typhimurium RO45, S. baildon 61-99, S. stanley H0558, S. enteritidis 15159, S. gaminara 02615, and S. anatum F4317. It should be noted that while the overlap between clusters is generally by nearest neighbor, isolates in the center of some clusters with larger standard errors have a broader range of statistical similarity. Examples of these are S. poona G91-1574, S. mbandaka RV1DHE, S. poona 953, and S. typhimurium DT104. Resistant isolates were observed to grow at the highest levels of antibiotic tested. Of the 24 isolates tested, 8 (33%) were resistant to Amp, 7 (29%) were resistant to Chl, and all 24 (100%) were shown to be resistant to Gm (Table 1). Isolates resistant to Amp, but not to Chl, include S. poona 348 and S. mbandaka 00916-1. One isolate (S. muenchen HERV2C) is resistant to Chl, but not to Amp. Six isolates were resistant to all the three antibiotics. They include S. poona G91-1574, S. poona 953, S. poona 418, S. st. paul FSIS-039, S. typhimurium DT104, and S. gaminara 02-615. w2 analysis shows that the resistant isolates are not significantly associated with any one quartile of radiation resistance. No correlation between antibiotic resistance and irradiation sensitivity was observed.
4. Discussion 3. Results Ionizing radiation effectively reduced the viable population of each of the isolates (Fig. 1). The radiation sensitivity was significantly variable among the various isolates (Table 1). The minimum D10 value observed was 0.181 kGy, while the maximum D10 value was 0.359 kGy, almost double that of the minimum value. An isolate’s source was not an indication of its D10 value. The D10 values are statistically similar to nearest neighbors, and form statistical clusters based on the ANCOVA results. To facilitate visualization of this clustering, the twenty-four Salmonella isolates were plotted with increasing D10 value (Fig. 2). Five major statistical clusters are observed in the data and are marked by ellipses in Fig. 2. The first cluster includes
The Salmonella isolates tested in this study had D10 values which ranged from 0.181 to 0.359 kGy. This range is consistent with previously published D10 values for Salmonella (Thayer and Rajkowski, 1999; Niemira, 2003). The statistical overlap and clustering of the D10 values is not apparently related to the foods from which the isolates were associated. The cluster with the highest D10 values contains isolates with a similar provenance as those found in the other clusters, such as isolates from cantaloupe, orange juice and tomato. A comprehensive understanding of why a given isolate may be more or less sensitive to irradiation than related isolates of the same pathogen has yet to be formulated (Thayer and Rajkowski, 1999; Niemira, 2003). It is likely that a survey of a larger number of isolates would result in a more linear progression of D10 values, with values that
ARTICLE IN PRESS 1084
B.A. Niemira et al. / Radiation Physics and Chemistry 75 (2006) 1080–1086
Fig. 1. Radiation sensitivity of twenty-four Salmonella clinical isolates in buffer. Isolates are divided in quartiles (A, B, C and D) by increasing radiation D10 values, six isolates per quartile.
bridge the gap seen among the statistical clusters observed among these 24 isolates. Antibiotic resistance among human pathogens is of increasing concern, particularly with regard to the potential for emergence of multi-drug-resistant strains (Golding and Matthews, 2004). Each of the 24 isolates examined in this study were resistant to Gm. Published surveys indicate that the prevalence of resistance to Gm can depend on the collection being examined, and can vary from year to year. Of 87 Salmonella isolates collected from turkeys in 1998–2002, 44.8% were resistant to Gm, a proportion which varied from 21.4% to 61.1%, depending on the year (Malik et al., 2003). In a survey of 390 Salmonella abattoir isolates taken from hog, beef and chicken carcasses, approximately 5% were determined to be resistant to Gm (Larkin et al., 2004). A study of 29 Salmonella isolates taken from turkey production facilities and turkey intestinal tracts showed 52% of isolates resistant to Gm (Nayak et al., 2004). These studies cite the use of antibiotics in an agricultural setting as a likely factor in
the emergence of single- or multi-drug-resistant pathogens. The 24 isolates in the present study were taken from a variety of sources, including produce, meat and juice products. A point of commonality among them is that each of these isolates was associated with a foodborne illness outbreak. For the purposes of the present study, the uniformity of resistance to Gm means that no conclusions may be drawn with regard to radiation sensitivity. The majority of the isolates which were resistant to Amp were also resistant to Chl, suggesting the possibility of a co-inheritance of resistance genes to these antibiotics, possibly on a single plasmid. Confirmation of this possibility would further investigation and characterization of the molecular basis for the antibiotic resistance exhibited by these isolates. The results presented herein lead to the conclusion that there is no relationship between radiation D10 value, and antibiotic resistance to Amp and Chl (Table 2). Antibiotic-resistant bacteria were not preferentially associated with either low- or high-D10 value isolates
ARTICLE IN PRESS B.A. Niemira et al. / Radiation Physics and Chemistry 75 (2006) 1080–1086
1085
Fig. 2. Salmonella isolates in order of increasing radiation D10 value including error bars. Statistical clusters are marked by ellipses on the figure.
Table 2 w2 results showing no correlation between D10 value and antibiotic resistance Quartile
1st 2nd 3rd 4th w2
Ampicillin
Gentamicin
Chloramphenicol
r
s
r
s
r
s
1 2 4 1
5 4 2 5
6 6 6 6
0 0 0 0
1 2 3 1
5 4 3 5
p ¼ 0:212
p ¼ 0:99
p ¼ 0:528
s, sensitivity to antibiotic; r, resistance to antibiotic.
for either Amp or Chl. The modes of action of Amp (cell wall synthesis) and Chl (protein synthesis) are not related to nucleic acid structure or function. The primary mode of action of ionizing radiation is via hydrogen and hydroxyl radical molecules resulting from the ionization of water molecules within the target organism. These radicals can disrupt membranes and interfere with the functioning of proteins, but the most significant target within the cell is DNA, where radicals are responsible for strand breakage (Niemira, 2003). The quinolone antibiotic Nal inhibits DNA synthesis in bacteria by interfering with the functioning of DNA
gyrase and topoisomerase IV (Georgopapadakou et al., 1987; Khadursky and Cozzarelli, 1998). NalR strains of E. coli O157:H7 and Salmonella have been recently shown to be more sensitive to irradiation than the NalS parent strains from which they were derived (Niemira, 2005a,b). In the E. coli O157:H7 study (Niemira, 2005a), NalR strains were also exposed to the antibiotic before, during and after irradiation with no change in survival or radiation sensitivity. The mechanism by which the induced NalR protects the cell from the antibiotic was therefore determined to be insensitive to the irradiation process. The NalR isolates used in that study were
ARTICLE IN PRESS 1086
B.A. Niemira et al. / Radiation Physics and Chemistry 75 (2006) 1080–1086
generated using unguided selection rather than by a transformation with a specific insertion of a DNA construct. The precise molecular nature of the Nal resistance in those isolates was therefore not characterized. Nevertheless, it was concluded that the significant differences in D10 values between NalR isolates and their NalS parent isolates would lead to an overestimate of the efficacy of ionizing radiation against E. coli 0157:H7 in studies using the NalR daughter isolates (Niemira, 2005a). In the present study, resistance to Amp or Chl was not associated with sensitivity to ionizing radiation. Therefore, bacterial resistance to AmpR or ChlR, rather than NalR, are suggested as more appropriate markers in irradiation studies of inoculated food products using Salmonella. Acknowledgments The authors would like to thank Dr. E. Solomon for providing Salmonella isolates, Dr. J. Buescher for valuable discussion, and Drs. L. Huang and C.-H. Liao for their thoughtful reviews of this manuscript. References Beuchat, L.R., 1995. Pathogenic microorganisms associated with fresh produce. J. Food Prot. 59, 204–216. Blackburn, C. de W., Davies, A.R., 1994. Development of antibiotic-resistant strains for the enumeration of foodborne pathogenic bacteria in stored foods. Int. J. Food Microbiol. 24, 125–136. Foley, D.M., Dufor, A., Rodriguez, L., Caporaso, F., Prakash, A., 2002. Reduction of Escherichia coli 0157:H7 in shredded iceberg lettuce by chlorination and gamma irradiation. Radiat. Phys. Chem. 63, 391–396. Gebreyes, W.A., Davies, P.R., Turkson, P., Morrow, W.E.M., Funk, J.A., Althier, C., Thakur, S., 2004. Characterization of antimicrobial-resistant phenotypes and genotypes among Salmonella enterica recovered from pigs on farms, from transport trucks, and from pigs after slaughter. J. Food Prot. 67, 698–705. Georgopapadakou, H., Dis, B.A., Angehrn, P., Wick, A., Olsen, G.L., 1987. Monocyclic and tricyclic analogs of quinolones: mechanisms of action. Antimicrob. Agents Chemother. 31, 614–616. Golding, S.S., Matthews, K.R., 2004. Intrinsic mechanism decreases susceptibility of Escherichia coli O157:H7 to multiple antibiotics. J. Food Prot. 67, 34–39. Horby, P.W., O’Brien, S.J., Adak, G.K., Graham, C., Hawker, J.I., Hunter, P., Lane, C., Lawson, A.J., Mitchell, R.T., Reacher, M.H., Threlfall, E.J., Ward, L.R., 2003. A national outbreak of multi-resistant Salmonella enterica
serovar Typhimurium definitive phage type (DT) 104 associated with consumption of lettuce. Epidemiol. Infect. 130, 169–178. Khadursky, A.B., Cozzarelli, N.R., 1998. The mechanism of inhibition of topoisomerase IV by quinolone antibacterials. J. Biol. Chem. 273, 27668–27677. Larkin, C., Poppe, C., McNab, B., McEwen, B., Mahdi, A., Odumeru, J., 2004. Antibiotic resistance of Salmonella isolated from hog, beef and chicken carcass samples from provincially inspected abattoirs in Ontario. J. Food Prot. 67, 448–455. Malik, Y.S., Olsen, K., Chander, Y., Goyal, S.M., 2003. Antimicrobial resistance in bacterial pathogens isolated from turkeys in Minnesota from 1998 to 2002. Int. J. Appl. Res. Vet. Med. 1 (3) http://www.jarvm.com/articles/ Vol1Iss3/Malik.htm (Accessed October 5, 2005). Nayak, R., Stewart, T., Wang, R.F., Lin, J., Cerniglia, C.E., Kenney, P.B., 2004. Genetic diversity and virulence gene determinants of antibiotic-resistant Salmonella isolated from preharvest turkey sources. Int. J. Food Microbiol. 91, 51–62. NCCLS (National Committee for Clinical Laboratory Standards), 2003. Methods for dilution antimicrobial susceptibility tests for bacteria that grow aerobically, approved standard, sixth ed., NCCLS document M7-A6, Wayne, PA, ISBN 1-56238-486-4. Niemira, B.A., 2003. Irradiation of fresh and minimally processed fruits, vegetables and juices. In: Novak, J.S., Sapers, G.M., Juneja, V.K. (Eds.), The Microbial Safety of Minimally Processed Foods. CRC Press, Boca Raton, FL, pp. 279–300. Niemira, B.A., 2005a. Nalidixic acid resistance increases sensitivity of Escherichia coli 0157:H7 to ionizing radiation in solution and on green leaf lettuce. J. Food Sci. 79 (2), M121–M124. Niemira, B.A., 2005b. Nalidixic acid resistance increases sensitivity of Salmonella to ionizing radiation in buffer and in orange juice. Abs P1-33. International Association for Food Protection Annual Meeting, Baltimore, MD. Ray, B., 1979. Methods to detect stressed microorganisms. J. Food Prot. 42, 346–355. Sivapalasingam, S., Friedman, C.R., Cohen, L., Tauxe, R.V., 2004. Fresh produce: a growing cause of outbreaks of foodborne illness in the United States, 1973 through 1997. J. Food Prot. 67, 2342–2353. Taormina, P.J., Beuchat, L.R., 1999. Comparison of chemical treatments to eliminate enterohemorrhagic Escherichia coli O157:H7 on alfalfa seeds. J. Food Prot. 62, 318–324. Tauxe, R., Kruse, H., Hedberg, C., Potter, M., Madden, J., Wachsmuth, K., 1997. Microbial hazards and emerging issues associated with produce: a preliminary report to the National Advisory Committee on Microbiological Criteria for Foods. J. Food Prot. 60, 1400–1408. Thayer, D.W., Rajkowski, K.T., 1999. Developments in irradiation of fresh fruits and vegetables. Food Technol. 53, 62–65.